man in a business suit

Developing Organizational Talent: Why data is key for maximizing returns

read time - icon

0 min read

May 25, 2022

Data trumps intuition 

In 2002, the general manager of the Oakland A’s, Billy Beane, forever changed how baseball teams are built. After losing some of their star players to free agency and having a mere fraction of the salary cap allotted to top echelon teams like the New York Yankees, Beane realized that if he wanted to assemble a competitive baseball team, he’d have to do so without top-tier talent.

His approach revolutionized the baseball world. It centered on using data – quantitative metrics for players’ performance – which allowed him to uncover value that was overlooked by everyone else. Cold, hard data, as it turns out, serves as a better guide to understanding a player’s value than scouting intuition. 

Playing organizational ‘Moneyball’

This story, popularized in the blockbuster movie Moneyball, is not all too dissimilar to building and managing an organization: companies are tasked with allocating scarce resources (talent) in a way that maximizes value to the company. Organizations can imitate Beane’s Moneyball strategy by using data to better map talent to value – but it’s not just about what data you collect. It’s about how you use it. 

Organizations now have the ability to collect data every bit as objective as RBIs (runs batted in). Key performance indicators (KPIs) give a high-level evaluation of a company’s success, and these can be broken down into ever more discrete tasks that comprise the productive work in an organization. Data-driven insights on individual employees can be gleaned from personal assessments or behavioral experiments that more accurately elicit their preferences, interests and attributes. 

When used effectively, data can create a better alignment between organizational roles and employees’ skills and interests so that everyone has an opportunity to thrive. 

The promise and perils of data-driven decisions

When it comes to allocating talent in an organization, a data-driven approach has two benefits. 

  1. Making better talent reallocation decisions
    By more accurately evaluating employees’ skills and clarifying the tasks and responsibilities for a given position, it is easier to identify which employees are best-suited for specific roles.
  2. Quicker mobilization of organizational talent
    Disruptions in the workforce can cause shortages in key skill areas or mission-critical roles. Having up-to-date performance metrics can highlight internal talent that is best suited to fill a vacancy or take over key responsibilities in the interim. And more agile reallocation practices have shown a proven benefit in business performance. A McKinsey report found organizations that were quick to reallocate talent were more than twice as likely to outperform competitors than slow reallocators.

chart showing frequency of talent acquisition review

But using a data-driven approach comes with risks. If the decision-making process or the data itself are biased, then the approach will not deliver its intended value. Getting talent reallocation decisions right requires a careful examination of the cognitive biases affecting our judgment–which may be more subtle than you think. 

An illustrative example: the Peter Principle

When it comes to allocating talent within an organization, conventional wisdom is not always the best guide. Take, for instance, the Peter Principle: it posits that in a hierarchical organization every employee will rise to the level of their incompetence.2

The logic here is that a top-performing employee will get promoted until they settle into a suboptimal position that doesn’t warrant their further promotion. Although the Peter Principle was originally intended as a satirical remark, for many organizations it hits a little too close to home. Organizations that commit this error assume that the most ‘talented’ or top performing employee is best suited for promotion. This may apply in some cases, but when a new role requires different skills and personality traits, employee performance often suffers. 

Data can shine light on organizational blind spots

Data is not helpful if it feeds into a decision-making process that relies on faulty assumptions. Data can give a more accurate indication of top performing employees, but it is not always the top performers that are the best candidates for rising the organizational ranks. 

To avoid these decision-making blindspots, companies can apply behavioral science insights to leverage their data more effectively. To better understand the biases impacting reallocation decisions, it helps to unpack the Peter Principles in more detail. 

What the Peter Principle gets wrong: The halo effect

Why do we assume that the top performing employee is the most suitable for promotion? One explanation is the halo effect, which occurs when our positive perceptions of someone in one area are misapplied to another area. It is due to the halo effect that we may find that the top performing employee in sales doesn’t make the best manager, nor does the all-star sports player make the best coach. 

halo effect illustration

What underlies this bias is a tendency to attribute the current or anticipated success of a person to their intrinsic characteristics. But what we fail to consider is that a person’s performance – whether good or bad – is also highly situational: success can have as much to do with one’s environment as it is to do with personal capability. In other words, an employee’s performance depends on both their talent and fit within an organizational context.

performance = intrinsic ability x environment

This distinction is crucial because it recognizes that talent isn’t always reflected by performance. There may be hidden talent in an organization that is unable to flourish in a role that they are not aligned with or where they are lacking support. 

The key for employers is to use data to delineate intrinsic capabilities from environmental enablers to better understand where employees deliver the most value.

In practice: Concrete steps to go beyond correlation

  1. Identifying employee performance gaps
    The first step in mapping talent to value is to identify gaps in employee performance. This can be measured by customized KPIs and OKRs (objectives and key results), which specify where an employee or team is struggling and where they excel. Furthermore, data generated from employee feedback surveys can indicate where employees are struggling and guide intervention strategies to best support them.
  2. Identifying causal factors
    The next step is using the data to gain insights into causal factors underlying a performance deficit. This can be done by leveraging longitudinal KPI and employee feedback data, which can be further processed by machine learning models to understand how environmental changes affect a desired outcome. A causal model can answer questions like ‘how does a dramatic shift in work volume impact employee stress levels?’ Or ‘how is employee engagement affected after implementing a new management system?’
  3. Putting insights into practice
    The goal of the causal analysis is to identify key factors that boost or detract from an employee’s performance. These insights can then be used to adapt their role and environment accordingly. Data can also be used to evaluate and iteratively improve interventions that address the root cause of performance gaps.

    By distinguishing between personal and environmental factors that impact performance, organizations can make more accurate talent reallocation decisions while creating more personalized roles that play to every employee’s strengths.

Mapping talent to value to beat the halo effect

Mapping talent to value is key for the success of an organization and data can be essential for making these decisions more accurate and agile. But data-driven decisions are still prone to cognitive biases like the halo effect. By having a more nuanced understanding of the causal factors that impact performance, companies can make more objective, evidence-based decisions that maximize the value of organizational talent.  

The Decision Lab is a research-oriented consultancy that uses behavioral science to advance social good. We work with some of the largest organizations in the world to spark change and tackle tough societal problems. Data-driven decision-making is key to eliminating bias in the workplace and maximizing the talent organizations already have on hand. If you'd like to tackle this together in your own workplace, contact us.

References

  1. Barriere, M., Owens, M., & Pobereskin, S. (2018). Linking talent to value. McKinsey & Company. Retrieved May 5, 2022, from https://www.mckinsey.com/business-functions/people-and-organizational-performance/our-insights/linking-talent-to-value 
  2. Peter, L. J., & Hull, R. (1970). The Peter principle. New York: Bantam.

About the Authors

Ryan McPhedrain

Ryan McPhedrain

Ryan poursuit actuellement son doctorat en neurosciences à l'Université McGill, en se concentrant sur les mécanismes moléculaires et cellulaires de la plasticité neuronale dans le cerveau en développement. Il s'intéresse principalement à l'application de cadres comportementaux pour guider les interventions visant à améliorer la santé mentale et le bien-être. Fervent défenseur des solutions fondées sur les données, il cherche à tirer parti de la science des données et des outils d'apprentissage automatique pour améliorer les résultats comportementaux dans les domaines de la santé et de la finance numériques. Il a également participé à des campagnes de sensibilisation aux sciences affiliées à McGill, en donnant des présentations sur des sujets liés aux neurosciences à des élèves du secondaire et en répondant à des questions sur les neurosciences posées par le public. Dans ses temps libres, Ryan aime lire un bon livre, pratiquer divers sports comme le hockey, le volley-ball et le tennis, ou simplement se perdre dans la nature.

Read Next

Notes illustration

Vous souhaitez savoir comment les sciences du comportement peuvent aider votre organisation ?